 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am I the only one who thinks that the confidence value in the focal blur
is not scaled very well? The default value (if I remember right, 0.9) is
not very good, and every time I want decent focal blur, I have to set it
to something like 0.9999 and I still get a bit of graininess. So 0.9999 seems
to be a minimum to get decent focal blur; 0.9 is not enough; all the values
below 0.9 are completely useless.
This may be too late, but I think that it should be scaled better. For
example 0.5 would be the default value and it would give the same quality as
the current 0.9, and the quality which you get currently with 0.9999 would
be got with something like 0.9 or 0.95 or something.
--
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):5;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Nieminen Mika wrote:
>
> Am I the only one who thinks that the confidence value in the focal blur
> is not scaled very well? The default value (if I remember right, 0.9) is
> not very good, and every time I want decent focal blur, I have to set it
> to something like 0.9999 and I still get a bit of graininess. So 0.9999 seems
> to be a minimum to get decent focal blur; 0.9 is not enough; all the values
> below 0.9 are completely useless.
> This may be too late, but I think that it should be scaled better. For
> example 0.5 would be the default value and it would give the same quality as
> the current 0.9, and the quality which you get currently with 0.9999 would
> be got with something like 0.9 or 0.95 or something.
>
> --
> main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
> ):5;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/
I have wondered about this myself in the past and blamed it on my lack
of experience with the process.
I think the same holds true for the extinction level setting used in
scattering media. An extinction level of one is horrible while a value
of 0.1 is much closer to what one might expect. It too seems to suffer
from a scaling problem... perhaps a misplaced decimal point in the
source code.
--
Ken Tyler
See my 700+ Povray and 3D Rendering and Raytracing Links at:
http://home.pacbell.net/tylereng/index.html
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ken wrote:
> I have wondered about this myself in the past and blamed it on my lack
> of experience with the process.
That is three of us.
> I think the same holds true for the extinction level setting used in
> scattering media. An extinction level of one is horrible while a value
> of 0.1 is much closer to what one might expect. It too seems to suffer
> from a scaling problem... perhaps a misplaced decimal point in the
> source code.
But I would guess it is not a mistake but a toss up between the
number the equation requires and the log of that number. For
example, a common analog volume control feels linear but the
amount it changes internally is logarithmic. If it were linear
lowest hearable to highest volume would be in the last ten
percent. (Not quite but that is the idea.) Since they have to
pick some rational scale make it a judgement call.
--
<blink>-------please--don't-----------------</blink>
http://www.giwersworld.org/artii/
http://www.giwersworld.org/artiii/
Finally up on 99/06/22 updated 07/13
Post a reply to this message
Attachments:
Download 'us-ascii' (1 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Matt Giwer wrote:
> But I would guess it is not a mistake but a toss up between the
> number the equation requires and the log of that number. For
> example, a common analog volume control feels linear but the
> amount it changes internally is logarithmic. If it were linear
> lowest hearable to highest volume would be in the last ten
> percent. (Not quite but that is the idea.) Since they have to
> pick some rational scale make it a judgement call.
Yes but the use of logarithmic pots for audio purposes has to do with
the physics of human hearing and not to do with my visual pleasure when
using media. It could easily be linear or log and it would not matter
if the scaling of the feature were inadaquate for my purposes.
--
Ken Tyler
See my 700+ Povray and 3D Rendering and Raytracing Links at:
http://home.pacbell.net/tylereng/index.html
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ken wrote:
> Yes but the use of logarithmic pots for audio purposes has to do with
> the physics of human hearing and not to do with my visual pleasure when
> using media. It could easily be linear or log and it would not matter
> if the scaling of the feature were inadaquate for my purposes.
Let me rephrase without appearing to post for the sake of
posting.
There are semi-log functions and quatro-log functions. At
deci-log it might as well be linear. At some point they have to
pick one number range. Phong is technically as wrong as this one
but the complaints have died down.
As I am constantly reminded, POV is not just for folks who
understand what we are talking about with linear and log.
If I see anything (still having the source code but not having
screwed up my courage to the sticking point to dive into it) I
see a mixture of log(0-1), log(0-10) and no log at all. Usually a
range of numbers is in the docs.
I will say two things about the docs. The docs are poor. The
docs are better than most programs over $500. I know of no higher
compliment for a free program.
As for this problem, when I try something new and it does not
work as I expected I treat it like an alpha tester. I try the
extremes first. 1000 and then .001 then go to .5. Hit the
extremes and the average. And honestly after much frustration I
usually discover I was doing something wrong in the first place.
There is an old trick cut by halves. In eight tries you are down
to one part in 64k. That will quickly show you where effective
changes are.
Post a reply to this message
Attachments:
Download 'us-ascii' (1 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Matt Giwer wrote:
Poorly stated but not easy to make scientifically correct
generalities. The 1 and 10 should have been +infinity for most I
have played with. From memory phong (not phong_size) says not
much difference over 5 which is quite a way from +infinity. But
even phong_size at 70 or 70000 is a long way from infinity. And
0-70 ia about the noticeable range of effect.
I presume readers familiar with evaluating equations know what I
am talking about. Can any of you help me explain?
A FAQ file of "these ranges produce the differences" would be
useful.
Post a reply to this message
Attachments:
Download 'us-ascii' (1 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Can you please stop sending your articles in MIME format and without any
vcards or anything? Just plain text. Thanks.
Here is quoted all the text of your article which is completely useless
and completely irrelevant. A total of 24 lines:
: This is a multi-part message in MIME format.
: --------------F5C7A739E1789723D1719CD4
: Content-Type: text/plain; charset=us-ascii
: Content-Transfer-Encoding: 7bit
: --------------F5C7A739E1789723D1719CD4
: Content-Type: text/x-vcard; charset=us-ascii;
: name="mgiwer.vcf"
: Content-Transfer-Encoding: 7bit
: Content-Description: Card for Matt Giwer
: Content-Disposition: attachment;
: filename="mgiwer.vcf"
: begin:vcard
: n:Giwer;Matt
: tel;cell:813-694-3998
: tel;work:813-972-2574
: x-mozilla-html:FALSE
: org:Images Incarnate
: version:2.1
: email;internet:mgi### [at] giwersworld org
: title:Owner
: adr;quoted-printable:;;14205 Nebraska Ave=0D=0A#27;Tampa;FL;33613;USA
: fn:Matt Giwer
: end:vcard
: --------------F5C7A739E1789723D1719CD4--
Why don't you make just a regular signature like anyone else?
--
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):5;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Nieminen Mika wrote:
> Can you please stop sending your articles in MIME format and without any
> vcards or anything? Just plain text. Thanks.
> Here is quoted all the text of your article which is completely useless
> and completely irrelevant. A total of 24 lines:
I look at my own messages and see no such thing. Why don't you
just adjust your reader not to show the message headers? Every
message has provision for at least 1k worth of header I think but
I would have to check the RFC to be sure.
> Why don't you make just a regular signature like anyone else?
I see plenty of Vcards around here. Why are you singling me out
for your attentions?
--
<blink>-------please--don't-----------------</blink>
http://www.giwersworld.org/artii/
http://www.giwersworld.org/artiii/
Finally up on 99/06/22 updated 07/13
Post a reply to this message
Attachments:
Download 'us-ascii' (1 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>Nieminen Mika wrote:
BTW TIN is a very primitive newsreader. You might consider
upgrading.
Post a reply to this message
Attachments:
Download 'us-ascii' (1 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Tue, 10 Aug 1999 07:34:29 -0400, Matt Giwer wrote:
>This is a multi-part message in MIME format.
>--------------130CED139398AB15E1C28502
>Content-Type: text/plain; charset=us-ascii
>Content-Transfer-Encoding: 7bit
^^^ Those aren't headers. They're MIME Multipart separators.
> I look at my own messages and see no such thing. Why don't you
>just adjust your reader not to show the message headers? Every
A) That's because you're using a web browser instead of a news reader.
B) Those aren't headers. Even if you turn off headers, the useless MIME
crap shows up. MIME is okay in binaries groups, because we expect it
there, but it is out of place in a text-only group.
I'm usually happy to just ignore it, but when people are wrong and then
get belligerent about their right to be wrong rather than solving the
problem, it starts to bother me.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
